By Mahnoor Faisal

It took me longer than I’d like to admit to realize that not all AI tools are created equal. Some excel at digging up information, while others shine at helping you make sense of it. I’ve always been very picky about the AI tools I use. When I find one that works well for me, I cling to it and obsessively explore every feature until I’m using it to its full potential. For me, this truly happened when I discovered Google was testing an AI-powered personalized research assistant in Labs called NotebookLM (codenamed Project Tailwind back then).

I’ve been using the tool since its experimental days, and for the longest time, I exclusively relied on NotebookLM. It was (and still frankly is) the perfect tool for me, both as a student and a journalist. It had everything I wanted in an AI tool, so I figured, “Why bother looking for more tools when this one already checks all the boxes?” That was one of the biggest mistakes I could’ve made.

A couple of weeks ago, I paired NotebookLM with Perplexity, and it changed everything. They made an incredibly powerful duo, and it made NotebookLM a lot more powerful. Then, I decided to pair it with multiple other tools, including ChatGPT. Then, I got a wild idea: what if I combined all three (NotebookLM, Perplexity, and ChatGPT) into a single workflow? That’s exactly what I decided to do, and it might just be the ultimate knowledge power combo I’ve been looking for.

Perplexity helps find reliable sources fast

Perplexity is an AI-powered search engine designed to reduce the hours you spend manually searching for information. Traditionally, you'd type a question into Google and then browse through the results to find what you're looking for. Perplexity does all that for you, and in my months of using it for studying, it’s yet to let me down in terms of reliability.

But when it comes to pairing it with NotebookLM and ChatGPT, I don’t really use it to search for information or get answers. Instead, I rely on it specifically to quickly track down credible sources depending on what I’m studying or researching. For instance, say I’m revising for an upcoming midterm.

I use NotebookLM for all my studying, but before I dive in, I need solid sources to upload to my notebooks. Though I already have lecture slides and my own notes, sometimes I want to supplement my notebook with up-to-date articles, tutorials, or official documentation. All I need to do is send a prompt like:

Find credible sources explaining the key principles of object-oriented programming, including encapsulation, inheritance, and polymorphism. Provide links to articles, tutorials, or papers that offer clear examples and explanations

Though Perplexity essentially answers my query, I instantly switch from the Answer tab to the Sources tab, which lists all the sources the AI used to find that answer. Now, all I have to do is import the sources it found into my NotebookLM notebook.

Sometimes, I skim through Perplexity’s answer to help me filter which sources to add to my notebook. However, given that Perplexity is considered one of the most reliable AI-powered search tools out there, I usually trust the sources it pulls up without much second-guessing. This way, I save hours I would’ve otherwise spent manually hunting down credible articles or papers.

NotebookLM then helps me make sense of everything

As I mentioned above, I don't use Perplexity's insights to actually understand the material. Though I have no doubt that what it finds is going to be great, Perplexity doesn't currently have the features that help me learn best. NotebookLM, on the other hand, does. For instance, I know I've retained a lot of information from random podcasts I've listened to before.

Thankfully, NotebookLM's Audio Overviews feature is perfect for that. It can analyze all my uploaded sources and, instead of just giving me a text-based summary (which it can), it can generate an easy-to-follow audio summary that I can listen to every time. I live an hour away from my current college, which makes my commute the perfect time to listen to these audio summaries. Instead of scrolling through my phone or staring at a screen, I can absorb key information hands-free, turning otherwise “dead time” into productive study sessions.

For the longest time, I only generated Audio Overviews of sources I already had, like my lecture slides and assignments I had worked on. Essentially, I already had some familiarity with the sources, which meant that I wasn’t always fully relying on the Audio Overviews to learn something completely new. Adding sources I find from Perplexity really helps with that.

NotebookLM also has other features like Mind Maps, which really come in handy with this particular workflow. Since I use a combination of my own curated sources and ones Perplexity finds me, but they’re both about thesame topic, Mind Maps help me see how different pieces of information connect and relate. This visual overview makes it easier to organize my thoughts and spot gaps in my understanding, so I can focus my studying where it matters most.

I then use ChatGPT to dig a bit deeper

Before I integrated ChatGPT into my NotebookLM-Perplexity workflow, I’d simply dump the sources Perplexity found into a NotebookLM notebook and then use NotebookLM to interact with it. Now, I use ChatGPT to expand even more on NotebookLM’s outputs. If you’re familiar with NotebookLM, you likely know its biggest benefit is its source-grounded responses.

What this essentially means is that it only pulls information from the documents you upload, which makes it highly reliable for your own research. However, when I’m preparing for exams or trying to learn something new, there are times when fresh questions pop up as a result of a response from NotebookLM I just read. Of course, those questions are not always directly answered in the sources I’ve uploaded. And since NotebookLM only draws from what I’ve provided, it can’t help me explore beyond that.

That’s where ChatGPT comes in. I paste the NotebookLM response that made me think of the question I have as-is, then ask ChatGPT to dive deeper or clarify the concept further. Sometimes, I ask it to provide real-world examples, alternative explanations, or even help me connect that idea to other topics I’m studying.

ChatGPT’s newest Study Mode is also a game-changer in this workflow. It’s designed to help learners like me engage with complex material through interactive explanations, quizzes, and step-by-step walkthroughs. When I feed it the NotebookLM outputs along with my questions, Study Mode can break down the content into manageable chunks and even test my understanding.

Perplexity, NotebookLM, and ChatGPT complete my workflow

I’ve paired NotebookLM with numerous AI tools before — Slack, Google Docs, Excel. You name it. But none have come close to matching the experience I get from combining these three.